CN116703682B - Government affair data platform based on deep learning - Google Patents
Government affair data platform based on deep learning Download PDFInfo
- Publication number
- CN116703682B CN116703682B CN202310988091.3A CN202310988091A CN116703682B CN 116703682 B CN116703682 B CN 116703682B CN 202310988091 A CN202310988091 A CN 202310988091A CN 116703682 B CN116703682 B CN 116703682B
- Authority
- CN
- China
- Prior art keywords
- government
- service
- entity
- government service
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013135 deep learning Methods 0.000 title claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000003062 neural network model Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims description 20
- 238000005295 random walk Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012804 iterative process Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Animal Behavior & Ethology (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a government affair data platform based on deep learning, and belongs to the technical field of online service. The government affair data platform based on deep learning comprises a government affair service method, wherein the method comprises the following steps of obtaining historical government affair service data; training based on historical government service data to obtain a first neural network model; acquiring new government service data of a user; outputting the difficulty of the government service corresponding to the new government service data in the first neural network model based on the new government service data; the method comprises the steps of obtaining a predicted government service sequence based on new government service data of a user, and obtaining the difficulty of government service corresponding to the new government service based on the predicted government service sequence; evaluating the capability of the service personnel to obtain a capability value of the service personnel; the government affair data platform based on deep learning provides an optimized government affair personnel service model, and improves user experience.
Description
Technical Field
The application belongs to the technical field of online service, and particularly relates to a government affair data platform based on deep learning.
Background
With the wave of the Internet plus technology, the online government service technology is more and more widely applied, and compared with the traditional window service and telephone customer service, the online government service system based on the Internet has the advantages of real-time access, automatic customer service distribution, high service efficiency, real-time monitoring of session data and the like.
At present, an online government service system still adopts a service personnel to serve a plurality of users in a service mode, and with the complexity of business and the refinement of division, one service personnel cannot answer all types of consultations. In the existing online government service system, a service switching mode is generally adopted to switch the session to another service person for service, and new problems are brought to influence the user experience. Meanwhile, since the user may not know what service needs to be handled by himself, a lot of time is consumed at the beginning of selecting the service type, and user experience is affected.
In the prior art, a data platform for interfacing the user and the service personnel can be provided, but government service is generally randomly distributed to idle service personnel, but the capability of the distributed service personnel and the difficulty of the user problem are generally not considered, and if the capability of the distributed service personnel is lower than the processing difficulty of the government service content, the service quality is easily degraded.
Disclosure of Invention
The application aims to solve the problems and provide the government affair data platform based on deep learning, which has a simple structure and a reasonable design.
The application realizes the above purpose through the following technical scheme:
the first aspect of the present application provides a government affair data service method, the method comprising the steps of,
step S102, historical government affair service data are obtained;
step S104, training and obtaining a first neural network model based on historical government affair service data;
step S106, obtaining new government service data of the user;
step S108, outputting the difficulty of the government service corresponding to the new government service data in the first neural network model based on the new government service data;
the method comprises the steps of obtaining a predicted government service sequence based on new government service data of a user, and obtaining the difficulty of government service corresponding to the new government service based on the predicted government service sequence;
step S110, evaluating the capability of service personnel to obtain a capability value of the service personnel;
step S112, service personnel with the matching capability value larger than the difficulty of the government service of the user serve the current user.
As a further optimization scheme of the application, historical government service data is extracted, entity identification and extraction are carried out, and an entity related to the problem is obtained; and training a graph neural network based on the knowledge graph of the government affair service.
As a further optimization scheme of the application, the knowledge graph of the government affair service comprises a problem entity and a user information entity, wherein the existence of the relation between the problem entity and the problem entity indicates that the two problem entities appear in the same government affair service, and the existence of the relation between the problem entity and the user information entity indicates that the user information entity and the problem entity appear in the same government affair service.
As a further optimization scheme of the application, in step S108, step S1081 extracts a problem entity and a user information entity from the current government service data, performs random walk on a knowledge graph with the user information entity and the problem entity as the center to generate a plurality of subgraphs, and splices the generated subgraphs to form an input graph.
As a further optimization scheme of the application, an input graph is input to a graph neural network, and the classification label of the graph neural network is the ID of the next adjacent problem entity of the problem entity; one entity of the input graph corresponds to one node, and edges between the nodes correspond to the association of the entities of the input graph; the corresponding node characteristics are obtained through word vector coding user information entities and problem entities, and a CBOW model (continuous word bag model) is adopted; the graphic neural network comprises a K-layer hidden layer and a full-connection layer.
As a further optimization scheme of the application, the calculation formula of the u-th layer of the graph neural network is as follows:； wherein ,/>Intermediate feature representing the ith node of the ith layer,/->Represents a set of nodes directly connected to node i, < >>Intermediate features representing the j-th node of the u-1 layer,/>Weight parameter representing the layer u, +.>Representing a sigmoid activation function; when u=1, _a->，/>Node characteristics representing the j-th node; the node characteristics output by the last layer of the graph neural network are input into a full-connection layer, and the full-connection layer outputs classification labels; the adjacent problem entities refer to adjacent problem entities after the problem entities are sequenced according to time in the same government service; the next adjacent problem entity of the problem entity refers to the adjacent problem entities ordered thereafter; the ID of a problem entity refers to its number in the problem entity library.
As a further optimization scheme of the application, step S1083, add the next adjacent problem entity corresponding to the last problem entity in the current government service data in step S1082 into the current government service data in step S1081 to update it, and then return to step S1081 until the next step is entered after the iterative execution is performed N-1 times.
As a further optimization scheme of the application, in step S1084, N problem entities of the current government service data obtained after step S1083 are extracted, a problem sequence is obtained according to the updated sequence, the problem sequence is input into a first neural network model, and the first neural network model outputs the difficulty of the current government service.
As a further optimization scheme of the present application, step S110, the formula for calculating the capability value of the service personnel is as follows:； wherein ,/>Representing the ability of the c-th attendantForce value, H, represents the total number of government services that the c-th attendant has performed,/->And the difficulty value of the j-th government service performed by the c-th service personnel is represented.
The second aspect of the application provides a government affair data service platform, which comprises a user side, a service side and a cloud server, wherein the user side is used for inputting government affair services and uploading the government affair services to the cloud server; the server side obtains the capability value of the service personnel and uploads the capability value to the cloud server; the cloud server obtains government service data based on government service input by a user terminal, obtains difficulty corresponding to the government service based on the government service data, matches corresponding service personnel based on the difficulty corresponding to the government service and the capability value of the service personnel, and establishes corresponding communication transmission between the service personnel and a user.
The application has the beneficial effects that: according to the application, through the consideration of the difficulty of the user problem and the capability of the service personnel, the corresponding matching is carried out, so that a better service effect is achieved; in addition, when facing a new government service, the application only accurately predicts the overall difficulty of the government service about to be performed by the user under the condition of knowing the first service problem and the user information, provides an optimized government personnel service model and improves the user experience.
Drawings
FIG. 1 is a schematic illustration of a government affair data service method according to the present application;
FIG. 2 is a schematic illustration of a method of random walk of the present application;
FIG. 3 is a schematic diagram of a method of the present application for obtaining difficulty based on government service data;
fig. 4 is a schematic structural diagram of a government data service platform according to the present application.
Detailed Description
The present application will be described in further detail with reference to the accompanying drawings, wherein it is to be understood that the following detailed description is for the purpose of further illustrating the application only and is not to be construed as limiting the scope of the application, as various insubstantial modifications and adaptations of the application to those skilled in the art can be made in light of the foregoing disclosure.
As shown in FIG. 1, a government affair data service method comprises the following steps, step S102, obtaining historical government affair service data; step S104, training and obtaining a first neural network model based on historical government affair service data; step S106, obtaining new government service data of the user; step S108, outputting the difficulty of the government service corresponding to the new government service data in the first neural network model based on the new government service data; step S110, evaluating the capability of service personnel to obtain a capability value of the service personnel; step S112, service personnel with the matching capability value larger than the difficulty of the government service of the user serve the current user.
And obtaining a predicted government service sequence based on new government service data of the user, and obtaining the difficulty of the government service corresponding to the new government service based on the predicted government service sequence.
Extracting historical government service data, and carrying out entity identification and extraction to obtain an entity related to the problem; and training a graph neural network based on the knowledge graph of the government affair service.
The knowledge graph of the government service comprises a problem entity and a user information entity, wherein the existence of the relation between the problem entity and the problem entity indicates that the two problem entities appear in the same government service, and the existence of the relation between the problem entity and the user information entity indicates that the user information entity and the problem entity appear in the same government service.
Wherein the problem entity comprises: policy and legal issues, administrative issues, participation in governance issues, government information disclosure issues, public service issues, etc.; the user information entity comprises: age, school, working years, marital status, etc.
A government service represents a service provided by a primary government platform for a user. For example, a user continuously inquires about the questions of the price, interest rate, school district, etc. in the services provided by the one-time government platform, and belongs to the services provided by the one-time government platform before the inquiry is finished.
Step S108 includes: as shown in fig. 3, in step S1081, a problem entity and a user information entity are extracted from current government service data, random walk is performed on a knowledge graph with the user information entity and the problem entity as the center to generate a plurality of sub-graphs, and the generated sub-graphs are spliced to form an input graph.
In one embodiment of the application, a random walk includes: as shown in fig. 2, step S81, initializing the number of steps of the walk to 1, wherein the entity currently walking is the entity of the walk center; step S82, randomly selecting one entity from the directly connected entities of the current walking entity, updating the selected entity to the current walking entity, and accumulating 1 for the walking steps; step S83, iteratively executing step S82 until the number of steps of the running reaches a set step threshold (the entity is not repeatedly selected in the iterative execution process, if no new entity is available for selection in a certain iterative process, the step is terminated in advance), and then recording all the selected entities in the subgraph; after M times of random walk, completing the process of generating a subgraph by taking an entity as a center; wherein M is a set super parameter, the default value is 10, the step number threshold is a set super parameter, and the default value is 10.
And sub-graph splicing, namely directly splicing a plurality of sub-graphs, and deleting repeated entities to obtain an input graph.
In step S1082, the input map is input to the graph neural network, and the classification label of the graph neural network is the ID of the next adjacent problem entity of the problem entity.
One entity of the input graph corresponds to one node, and edges between the nodes correspond to the association of the entities of the input graph; corresponding node characteristics are obtained through word vector coding user information entities and problem entities, and the word vector coding adopts a CBOW model (continuous word bag model).
The graphic neural network comprises a K-layer hidden layer and a full-connection layer.
The calculation formula of the u layer of the graph neural network is as follows:； wherein ,intermediate feature representing the ith node of the ith layer,/->Represents a set of nodes directly connected to node i, < >>Intermediate features representing the j-th node of the u-1 layer,/>Weight parameter representing the layer u, +.>Representing a sigmoid activation function; when u=1, _a->，/>Representing node characteristics of the j-th node.
The node characteristics output by the last layer of the graph neural network are input into a full-connection layer, and the full-connection layer outputs classification labels.
The adjacent problem entities refer to adjacent problem entities after the problem entities are sequenced according to time in the same government service; the next adjacent problem entity of the problem entity refers to the next adjacent problem entity ordered thereafter.
The ID of a problem entity refers to its number in the problem entity library.
And (3) inputting an input diagram generated by the historical government service data during training, and obtaining the classification labels from the problem entity sequences in the historical government service data.
Step S1083, adding the next adjacent problem entity corresponding to the last problem entity in the current government service data in step S1082 into the current government service data in step S1081 to update the next adjacent problem entity, and returning to step S1081 until the next step is entered after the iteration is performed for N-1 times.
Step S1084, extracting N problem entities of the current government service data obtained after the step S1083, arranging according to the updated sequence to obtain a problem sequence, inputting the problem sequence into a first neural network model, and outputting the current government service difficulty by the first neural network model.
The first neural network model is a cyclic neural network, the cyclic neural network inputs the t-th problem entity of the problem sequence at the t-th time step, the output of the last time step is connected to a classifier, and the classification label of the classifier represents the difficulty of government service.
The first neural network model is a multi-layer perceptron, problem entities in the problem sequence are spliced after vectorization, the problem entities are spliced into a vector and then input into the multi-layer perceptron, the output of the multi-layer perceptron is connected to a classifier, and the classification label of the classifier represents the difficulty of government service.
In one embodiment of the application, the difficulty of the government service is defined in grades, and is divided into three grades, namely low, medium and high, respectively assigned with 1, 2 and 3, and assigned with the purpose of counting the capability of service personnel.
In one embodiment of the application, the difficulty of the government service is defined by a score, and the value range of the score isThe value range mean is discretized into ten point values, corresponding to ten class labels, respectively.
In step S110, the formula for calculating the capability value of the attendant is as follows:； wherein ,/>Representing the capability value of the c-th attendant, H representing the total number of government services that the c-th attendant has performed,/for the c-th attendant>Represents the jth by the c-th attendantDifficulty value of personal government service.
As shown in fig. 4, a government affair data service platform is provided, which comprises a user side, a service side and a cloud server, wherein the user side is used for inputting government affair services and uploading the government affair services to the cloud server; the server side obtains the capability value of the service personnel and uploads the capability value to the cloud server; the cloud server obtains government service data based on government service input by a user terminal, obtains difficulty corresponding to the government service based on the government service data, matches corresponding service personnel based on the difficulty corresponding to the government service and the capability value of the service personnel, and establishes corresponding communication transmission between the service personnel and a user.
Generally, when a user connects with a government platform, a problem is firstly asked out, however, the traditional processing mode only carries out data extraction through one problem data, judges the type of the problem to match with corresponding service personnel, the problem is a difficult distinction, some problems are only flow problems, the problems are very simple for the service personnel, but some problems possibly relate to a plurality of government departments, a certain difficulty exists, the difficulty of the problem cannot be ignored to randomly distribute the corresponding service personnel to carry out client connection, meanwhile, the first problem is always only the beginning, and based on personal information of the user, the problem can be predicted to extend other problem sequences, namely the overall difficulty of the government service is obtained through the problem sequences; not only is the application considered, the ability situation of service personnel is also considered, and the experience of users can be improved for experienced service personnel to process government service with higher difficulty; the application can accurately predict the overall difficulty of the government service to be performed by the user only under the condition of knowing the first service problem and the user information.
The embodiment of the application also discloses electronic equipment. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program capable of being loaded and executed by the processor to implement any of the methods described above.
The electronic device may be an electronic device such as a desktop computer, a notebook computer, or a cloud server, and the electronic device includes, but is not limited to, a processor and a memory, for example, the electronic device may further include an input/output device, a network access device, a bus, and the like.
A processor in the present application may include one or more processing cores. The processor performs the various functions of the application and processes the data by executing or executing instructions, programs, code sets, or instruction sets stored in memory, calling data stored in memory. The processor may be at least one of an application specific integrated circuit (Application SpecificIntegrated Circuit, ASIC), a digital signal processor (Digital Signal Processor, DSP), a digital signal processing device (Digital Signal Processing Device, DSPD), a programmable logic device (ProgrammableLogic Device, PLD), a field programmable gate array (Field Programmable Gate Array, FPGA), a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, and a microprocessor. It will be appreciated that the electronics for implementing the above-described processor functions may be other for different devices, and embodiments of the present application are not particularly limited.
The memory may be an internal storage unit of the electronic device, for example, a hard disk or a memory of the electronic device, or may be an external storage device of the electronic device, for example, a plug-in hard disk, a Smart Memory Card (SMC), a secure digital card (SD), or a flash memory card (FC) provided on the electronic device, or the like, and may be a combination of the internal storage unit of the electronic device and the external storage device, where the memory is used to store a computer program and other programs and data required by the electronic device, and the memory may be used to temporarily store data that has been output or is to be output, which is not limited by the present application.
The embodiment of the application also discloses a computer readable storage medium. A computer readable storage medium storing a computer program capable of being loaded by a processor and executing any one of the methods described above.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium and which, when executed, may comprise the steps of the above-described embodiments of the methods. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The above embodiments are not intended to limit the scope of the present application, so: all equivalent changes according to the method and principle of the present application should be covered by the protection scope of the present application.
Claims (2)
1. A government affair data platform based on deep learning is characterized in that the platform comprises a government affair data service method, the method comprises the following steps,
step S102, historical government affair service data are obtained;
step S104, training and obtaining a first neural network model based on historical government affair service data;
step S106, obtaining new government service data of the user;
step S108, outputting the difficulty of the government service corresponding to the new government service data in the first neural network model based on the new government service data;
the method comprises the steps of obtaining a predicted government service sequence based on new government service data of a user, and obtaining the difficulty of government service corresponding to the new government service based on the predicted government service sequence;
step S110, evaluating the capability of service personnel to obtain a capability value of the service personnel;
step S112, service personnel with the matching capability value larger than the difficulty of government service of the user serve the current user;
extracting historical government service data, and carrying out entity identification and extraction to obtain an entity related to the problem; training a graph neural network based on a knowledge graph of government affair service;
the knowledge graph of the government service comprises a problem entity and a user information entity, wherein the existence of the relation between the problem entity and the problem entity indicates that two problem entities appear in the same government service, and the existence of the relation between the problem entity and the user information entity indicates that the user information entity and the problem entity appear in the same government service;
in step S108, it includes:
step S1081, extracting a problem entity and a user information entity from current government service data, performing random walk on a knowledge graph by taking the user information entity and the problem entity as centers to generate a plurality of subgraphs, and splicing the generated subgraphs to form an input graph;
a random walk includes: step S81, initializing the number of the walking steps to be 1, wherein the entity of the current walking is the entity of the walking center; step S82, randomly selecting one entity from the directly connected entities of the current walking entity, updating the selected entity to the current walking entity, and accumulating 1 for the walking steps; step S83, iteratively executing step S82 until the number of steps of the walking reaches a set step threshold; the method comprises the steps of selecting an entity repeatedly in the iterative execution process, and terminating the step in advance if no new entity is available for selection in a certain iterative process; then recording all selected entities in the subgraph; after M times of random walk, completing the process of generating a subgraph by taking an entity as a center; wherein M is a set super parameter, the default value is 10, the step number threshold is a set super parameter, and the default value is 10;
step S1082, inputting the input graph into the graph neural network, wherein the classification label of the graph neural network is the ID of the next adjacent problem entity of the problem entity; one entity of the input graph corresponds to one node, and edges between the nodes correspond to the association of the entities of the input graph; corresponding node characteristics are obtained through word vector coding user information entities and problem entities;
the graphic neural network comprises a K-layer hidden layer and a full-connection layer;
the calculation formula of the u layer of the graph neural network is as follows:
；
wherein ,intermediate feature representing the ith node of the ith layer,/->Represents a set of nodes directly connected to node i, < >>Intermediate features representing the j-th node of the u-1 layer,/>Weight parameter representing the layer u, +.>Representing a sigmoid activation function;
when u=1, the number of the cells,，/>node characteristics representing the j-th node;
the node characteristics output by the last layer of the graph neural network are input into a full-connection layer, and the full-connection layer outputs classification labels;
the adjacent problem entities refer to adjacent problem entities after the problem entities are sequenced according to time in the same government service;
the next adjacent problem entity of the problem entity refers to the adjacent problem entities ordered thereafter;
the ID of the problem entity refers to the number of the problem entity in the problem entity library;
step S1083, adding the next adjacent problem entity corresponding to the last problem entity in the current government service data in step S1082 into the current government service data in step S1081 to update the next adjacent problem entity, and returning to step S1081 until the next step is entered after the iteration is performed for N-1 times;
step S1084, extracting N problem entities of the current government service data obtained after the step S1083, arranging according to the updated sequence to obtain a problem sequence, inputting the problem sequence into a first neural network model, and outputting the current government service difficulty by the first neural network model;
in step S110, the formula for calculating the capability value of the attendant is as follows:
；
wherein ,representing the capability value of the c-th attendant, H representing the total number of government services that the c-th attendant has performed,/for the c-th attendant>And the difficulty value of the j-th government service performed by the c-th service personnel is represented.
2. The deep learning-based government affair data platform according to claim 1, wherein: the platform also comprises a user side, a service side and a cloud server, wherein the user side is used for inputting government service and uploading the government service to the cloud server;
the server side obtains the capability value of the service personnel and uploads the capability value to the cloud server;
the cloud server obtains government service data based on government service input by a user terminal, obtains difficulty corresponding to the government service based on the government service data, matches corresponding service personnel based on the difficulty corresponding to the government service and the capability value of the service personnel, and establishes corresponding communication transmission between the service personnel and a user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310988091.3A CN116703682B (en) | 2023-08-08 | 2023-08-08 | Government affair data platform based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310988091.3A CN116703682B (en) | 2023-08-08 | 2023-08-08 | Government affair data platform based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116703682A CN116703682A (en) | 2023-09-05 |
| CN116703682B true CN116703682B (en) | 2023-10-31 |
Family
ID=87841950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310988091.3A Active CN116703682B (en) | 2023-08-08 | 2023-08-08 | Government affair data platform based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116703682B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117372206A (en) * | 2023-12-08 | 2024-01-09 | 烟台云朵软件有限公司 | Meta-universe unified service integration method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429292A (en) * | 2019-10-22 | 2020-07-17 | 北京芝麻保科技有限公司 | Intelligent matching system for insurance industry service personnel and service objects |
| CN112364172A (en) * | 2020-10-16 | 2021-02-12 | 上海晏鼠计算机技术股份有限公司 | Method for constructing knowledge graph in government official document field |
| CN113822517A (en) * | 2021-02-03 | 2021-12-21 | 浪潮云信息技术股份公司 | Case division method and device based on capability matching |
| CN115099629A (en) * | 2022-06-28 | 2022-09-23 | 中国银行股份有限公司 | Customer service demand evaluation distribution method and system, electronic device and storage medium |
| CN115880120A (en) * | 2023-02-24 | 2023-03-31 | 江西微博科技有限公司 | Online government affair service system and service method |
| CN115905538A (en) * | 2022-12-13 | 2023-04-04 | 浙江大华技术股份有限公司 | Event multi-label classification method, device, equipment and medium based on knowledge graph |
| WO2023125779A1 (en) * | 2021-12-31 | 2023-07-06 | 中兴通讯股份有限公司 | Model training method, customer service agent matching method, server, and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10643140B2 (en) * | 2014-05-01 | 2020-05-05 | International Business Machines Corporation | Method, system and computer program product for automating expertise management using social and enterprise data |
| US20210216849A1 (en) * | 2020-01-14 | 2021-07-15 | Vmware, Inc. | Neural-network-based methods and systems that generate forecasts from time-series data |
-
2023
- 2023-08-08 CN CN202310988091.3A patent/CN116703682B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429292A (en) * | 2019-10-22 | 2020-07-17 | 北京芝麻保科技有限公司 | Intelligent matching system for insurance industry service personnel and service objects |
| CN112364172A (en) * | 2020-10-16 | 2021-02-12 | 上海晏鼠计算机技术股份有限公司 | Method for constructing knowledge graph in government official document field |
| CN113822517A (en) * | 2021-02-03 | 2021-12-21 | 浪潮云信息技术股份公司 | Case division method and device based on capability matching |
| WO2023125779A1 (en) * | 2021-12-31 | 2023-07-06 | 中兴通讯股份有限公司 | Model training method, customer service agent matching method, server, and storage medium |
| CN115099629A (en) * | 2022-06-28 | 2022-09-23 | 中国银行股份有限公司 | Customer service demand evaluation distribution method and system, electronic device and storage medium |
| CN115905538A (en) * | 2022-12-13 | 2023-04-04 | 浙江大华技术股份有限公司 | Event multi-label classification method, device, equipment and medium based on knowledge graph |
| CN115880120A (en) * | 2023-02-24 | 2023-03-31 | 江西微博科技有限公司 | Online government affair service system and service method |
Non-Patent Citations (1)
| Title |
|---|
| 问答***中复杂问题分解方法研究综述;冯 钧;《计算机工程与应用》;第58卷(第17期);第23-31页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116703682A (en) | 2023-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110598206B (en) | Text semantic recognition method and device, computer equipment and storage medium | |
| CN110363449B (en) | Risk identification method, device and system | |
| CN110032632A (en) | Intelligent customer service answering method, device and storage medium based on text similarity | |
| CN109345399A (en) | Claims Resolution methods of risk assessment, device, computer equipment and storage medium | |
| WO2020224106A1 (en) | Text classification method and system based on neural network, and computer device | |
| CN110458324B (en) | Method and device for calculating risk probability and computer equipment | |
| WO2020038100A1 (en) | Feature relationship recommendation method and apparatus, computing device and storage medium | |
| CN112989059A (en) | Method and device for identifying potential customer, equipment and readable computer storage medium | |
| CN116703682B (en) | Government affair data platform based on deep learning | |
| CN113254649B (en) | Training method of sensitive content recognition model, text recognition method and related device | |
| CN115018656B (en) | Risk identification method, and training method, device and equipment of risk identification model | |
| CN114780746A (en) | Knowledge graph-based document retrieval method and related equipment thereof | |
| CN112258250A (en) | Target user identification method and device based on network hotspot and computer equipment | |
| CN114238656A (en) | Reinforced learning-based affair atlas completion method and related equipment thereof | |
| CN116993185B (en) | Time sequence prediction method, device, equipment and storage medium | |
| WO2021000411A1 (en) | Neural network-based document classification method and apparatus, and device and storage medium | |
| CN115952438A (en) | Social platform user attribute prediction method and system, mobile device and storage medium | |
| CN115600818A (en) | Multi-dimensional scoring method and device, electronic equipment and storage medium | |
| CN114513578A (en) | Outbound method, device, computer equipment and storage medium | |
| CN113722584A (en) | Task pushing method and device and storage medium | |
| CN111291196B (en) | Knowledge graph perfecting method and device, and data processing method and device | |
| CN113468604A (en) | Big data privacy information analysis method and system based on artificial intelligence | |
| CN112507170A (en) | Data asset directory construction method based on intelligent decision and related equipment thereof | |
| CN113792163B (en) | Multimedia recommendation method and device, electronic equipment and storage medium | |
| CN116955560B (en) | Data processing method and system based on thinking chain and knowledge graph |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |